[Lldb-commits] [lldb] [llvm] [AArch64] move extension information into tablgen (PR #90987)

[Tomas Matheson via lldb-commits]

================
@@ -426,30 +532,40 @@ def FeatureSpecRestrict : SubtargetFeature<"specrestrict", "HasSpecRestrict",
   "true", "Enable architectural speculation restriction (FEAT_CSV2_2)">;
 
 def FeatureSB : Extension<"sb", "SB",
-  "Enable v8.5 Speculation Barrier (FEAT_SB)" >;
+  "Enable v8.5 Speculation Barrier (FEAT_SB)", [],
+  "FEAT_SB", "+sb", 470>;
 
 def FeatureSSBS : Extension<"ssbs", "SSBS",
-  "Enable Speculative Store Bypass Safe bit (FEAT_SSBS, FEAT_SSBS2)" >;
+  "Enable Speculative Store Bypass Safe bit (FEAT_SSBS, FEAT_SSBS2)", [],
+  "FEAT_SSBS", "", 490>;
 
 def FeaturePredRes : Extension<"predres", "PredRes",
-  "Enable v8.5a execution and data prediction invalidation instructions (FEAT_SPECRES)" >;
+  "Enable v8.5a execution and data prediction invalidation instructions (FEAT_SPECRES)", [],
+  "FEAT_PREDRES", "+predres", 480>;
 
-def FeatureCacheDeepPersist : Extension<"ccdp", "CCDP",
+def FeatureCacheDeepPersist : SubtargetFeature<"ccdp", "CCDP", "true",
     "Enable v8.5 Cache Clean to Point of Deep Persistence (FEAT_DPB2)" >;
 
+// NOTE: BTI now has posfeat/negfeat, which it didn't before
+let ArchExtKindSpelling = "AEK_NONE" in
 def FeatureBranchTargetId : Extension<"bti", "BTI",
-    "Enable Branch Target Identification (FEAT_BTI)" >;
+    "Enable Branch Target Identification (FEAT_BTI)", [],
+    "FEAT_BTI", "+bti", 510>;
 
+let ArchExtKindSpelling = "AEK_RAND", MArchName = "rng" in
 def FeatureRandGen : Extension<"rand", "RandGen",
-    "Enable Random Number generation instructions (FEAT_RNG)" >;
+    "Enable Random Number generation instructions (FEAT_RNG)", [],
+    "FEAT_RNG", "+rand", 10>;
 
+let MArchName = "memtag" in
 def FeatureMTE : Extension<"mte", "MTE",
-    "Enable Memory Tagging Extension (FEAT_MTE, FEAT_MTE2)" >;
+    "Enable Memory Tagging Extension (FEAT_MTE, FEAT_MTE2)", [],
+    "FEAT_MEMTAG", "", 440>;
----------------
tmatheson-arm wrote:

Good point. There should be no behaviour change here caused by this patch. I've added a comment in f1d287a75139bd8a09ee9d596b10aa77a2b672b7 explaining that `"memtag"` means different things for FMV and non-FMV extensions. I've left the `SubtargetFeature` description as-is because I think it is most likely to be used for printing command line help in the future, and the FMV difference will not be relevant there.


https://github.com/llvm/llvm-project/pull/90987